1907.] 



Wheeler, Fungus-growiny Ants of North America. 



■35 



that overlying them and contained huge fungus-gardens on their flattened 

 floors. These gardens were 10-15 cm. high, of a yellowish color below 

 and made up very largely of triturated sorghum leaves. Above they were 

 bluish or greenish gray and this was the only portion that was permeated 

 and covered with the living mycelium, the lower portions having lost their 

 fungus-nourishing substances. The large amount of this exhausted leaf- 

 pulp still retained in the chambers, showed that Atta texana must differ 

 from some of the tropical species of this genus, v.hich carry it to the surface 



Fig. 11. Large Atta texana nest on the right bank of Barton Creek near .\ustin, Texas. 

 (Photograph by Messrs. Brues and Melander.) 



and eject it from the craters. The Texan species sim])ly keeps on building 

 up its gardens till they reach a consideral)le thickness while the mycelium 

 retreats to the more nutritive superficial layer. ]Many of the gardens in 

 the nest under discussion contained worker larvse and pu])a' in al)undance, 

 but no sexual forms, either mature or immature. Both in this and in the 

 previously described nest I found many specimens of a little myrmecophilous 

 cockroach, of which I shall have more to say in the .sequel. Although the 



